Dataset Information

Deep-learning-enabled antibiotic discovery through molecular de-extinction.

ABSTRACT: Molecular de-extinction aims at resurrecting molecules to solve antibiotic resistance and other present-day biological and biomedical problems. Here we show that deep learning can be used to mine the proteomes of all available extinct organisms for the discovery of antibiotic peptides. We trained ensembles of deep-learning models consisting of a peptide-sequence encoder coupled with neural networks for the prediction of antimicrobial activity and used it to mine 10,311,899 peptides. The models predicted 37,176 sequences with broad-spectrum antimicrobial activity, 11,035 of which were not found in extant organisms. We synthesized 69 peptides and experimentally confirmed their activity against bacterial pathogens. Most peptides killed bacteria by depolarizing their cytoplasmic membrane, contrary to known antimicrobial peptides, which tend to target the outer membrane. Notably, lead compounds (including mammuthusin-2 from the woolly mammoth, elephasin-2 from the straight-tusked elephant, hydrodamin-1 from the ancient sea cow, mylodonin-2 from the giant sloth and megalocerin-1 from the extinct giant elk) showed anti-infective activity in mice with skin abscess or thigh infections. Molecular de-extinction aided by deep learning may accelerate the discovery of therapeutic molecules.

SUBMITTER: Wan F

PROVIDER: S-EPMC11310081 | biostudies-literature | 2024 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Deep-learning-enabled antibiotic discovery through molecular de-extinction.

Wan Fangping F Torres Marcelo D T MDT Peng Jacqueline J de la Fuente-Nunez Cesar C

Nature biomedical engineering 20240611 7

Molecular de-extinction aims at resurrecting molecules to solve antibiotic resistance and other present-day biological and biomedical problems. Here we show that deep learning can be used to mine the proteomes of all available extinct organisms for the discovery of antibiotic peptides. We trained ensembles of deep-learning models consisting of a peptide-sequence encoder coupled with neural networks for the prediction of antimicrobial activity and used it to mine 10,311,899 peptides. The models p ...[more]

PMID: 38862735

Similar Datasets

Project description:Extinction learning refers to the phenomenon that a previously learned response to an environmental stimulus, for example the expression of an aversive behavior upon exposure to a specific context, is reduced when the stimulus is repeatedly presented in the absence of a previously paired aversive event. Extinction of fear memories has been implicated with the treatment of anxiety disease but the molecular processes that underlie fear extinctionare only beginning to emerge. Here we show that fear extinction initiates up-regulation of hippocampal insulin-growth factor 2 (Igf2) and down-regulation of insulin-growth factor binding protein 7 (Igfbp7). In line with this observation we demonstrate that IGF2 facilitates fear extinction, while IGFBP7 impairs fear extinction in an IGF2-dependent manner. Furthermore, we identify one cellular substrate of altered IGF2-signaling during fear extinction. To this end we show that fear extinction-induced IGF2/IGFBP7-signaling promotes the survival of 17-19 day-old newborn hippocampal neurons. In conclusion, our data suggests that therapeutic strategies that enhance IGF2-signaling and adult neurogenesis might be suitable to treat disease linked to excessive fear memory. We employed mice to investigate fear extinction in the hippocampus-dependent contextual fear conditioning paradigm. To this end, male C57BL/6J mice were exposed to the fear conditioning box (context) followed by an electric foot-shock which elicits the acquisition of conditioned contextual fear. For extinction training animals were repeatedly reexposed to the conditioned context on consecutive days (24h interval) without receiving the footshockagain (extinction trial, E). This procedure eventually results in the decline of the aversive freezing behavior. Mice that were exposed to the conditioning context without receiving fear conditioning training served as control groups. To gain a better understanding of the molecular processes underlying fear extinction we performed a genome-wide analysis of the hippocampal transcriptome during fear extinction. In the employed paradigm fear extinction is a gradual process. To capture the longitudinal course of fear extinction we decided to perform hippocampal microarray analysis at two time points: (1) After the first extinction trial (E1) when animals display high levels of aversive freezing behavior and (2) at the extinction trial on which the freezing behavior was significantly reduced when compared to E1. This extinction trial, in the case of this experiment E5, we termed “extinction trial low freezing” (ELF). Mice that were exposed to the conditioning context without receiving fear conditioning training served as control groups (3). For all three groups we hybridized 5 samples (biological replicates).

Dataset Information

Deep-learning-enabled antibiotic discovery through molecular de-extinction.

Publications

Deep-learning-enabled antibiotic discovery through molecular de-extinction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets